How to measure the input impedance of a laser diode/LED using a VNA?

Question

It would be nice to have some advice on an issue I'm having with a laser diode. I have a DFB laser diode that has no kind of series resistor connected to it in the package. I want to make a driver for it that does high-frequency modulation up to 100 MHz (square pulses) or so and want to know the frequency response including possible reflections. The diode is on-chip and is wire-bonded to 50-ohm traces on a PCB. The wire bonds are shorter than a centimeter.

At this point, I want to measure the input impedance and also the S11 parameter of this system to better understand it.

I know that the typical LED/laser diode has two regions where the impedance is vastly different. Therefore I believe I need to bias the diode first to the operating region and then set an appropriate power on the VNA so that I stay in this region and do the characterization.

What bugs me is that I know the photon output of a laser diode is drawn against the current flowing through it (I have this plot.) Since it is a current driven system, does it make sense to use a usual VNA? My plan is to use a bias T to give a DC bias current that puts the diode in the operating regime and then connect the VNA to the RF port of the bias T.

Is this the right way to characterize the frequency response of the laser diode? Should I include the bias tee in the calibration process?

Any help is appreciated.

Answer 1

Since it is a current driven system, does it make sense to use a usual VNA?

It won't cause any problems for the VNA that the laser has very low differential impedance at the operating point.

My plan is to use a bias T to give a DC bias current that puts the diode in the operating regime and then connect the VNA to the RF port of the bias T.

Yes, this is a perfectly valid way to do the test.

However, as Neil mentioned in comments, if you're planning to drive the laser with a high modulation depth, knowing the small-signal behavior at one or the other operating point won't be very helpful.

You might find that adding a few (10-40) ohms series resistance between the end of the feed line and the laser itself will reduce ringing in the optical signal (but of course it will require a higher output voltage from your current source).

You will also want to not use the fastest possible edge rates from your pulsed current source. If you only need 1 ns rise and fall times, don't use a source with 100 ps rise/fall times (or if you do, filter the the signal to increase the rise/fall times).